Measure spread with an explicit data model
Calculate standard deviation for a list of observations, with the population or sample convention chosen explicitly.
How to use this calculator
- Choose Population for the entire dataset of interest or Sample for an estimate from sampled observations.
- Enter comma- or whitespace-separated signed decimal observations; no grouping commas, empty entries or expressions.
- Use at least one observation for Population and two for Sample; compare variance and standard deviation with their different units.
Methodology and assumptions
For n observations, mean μ=Σx/n. Population variance is Σ(x−μ)²/n; sample variance is Σ(x−μ)²/(n−1). Standard deviation is the square root of the selected variance. The implementation scales accepted decimals to exact integers and sums exact centered deviations, avoiding cancellation-prone floating-point Σx²−(Σx)²/n.
Precision and rounding
Exact rational mean/variance and exact radical SD are retained. Decimal summaries use 12 significant figures; approximate square roots are labeled. Variance has squared input units; SD has input units.
Worked example
For 2,4,4,4,5,5,7,9 the mean is 40/8=5. Squared deviations are 9,1,1,1,0,0,4,16 and sum to 32. Population variance=32/8=4 and SD=2. Sample variance=32/7 and sample SD=√(32/7)≈2.13808993530.
Reference examples are specification evidence only; they do not claim that this interface or calculation engine has been tested.
Interpreting the result
Standard deviation is expressed in the same units as the input data and summarizes spread around the mean. It does not establish a normal distribution or explain why observations vary.
Unweighted finite list only, with at most 10,000 observations, 700,000 raw characters, 12 decimal places per value and the field’s magnitude bound. No missing-value imputation, grouped frequency data or inference of a normal distribution. Identical observations have zero spread.
Common mistakes
- Choose sample or population mode intentionally.
- Keep all observations in the same unit.
- Repeated values are separate observations when present in the dataset.
- A small standard deviation does not imply that every value is close to the mean.
Frequently asked questions
When should I choose population mode?
Use it when the entered observations are the complete group of interest under that model.
When should I choose sample mode?
Use it when the observations are treated as a sample and the specified sample formula is appropriate.
Does standard deviation use the same unit as the data?
Yes, unlike variance, it is reported in the original data unit.
Does it prove data are normally distributed?
No. Standard deviation alone does not describe a complete distribution.
Are comma-separated numbers thousands-formatted?
No. Separators divide individual observations.
Why is sample variance larger for the same list?
Its denominator is n−1 rather than n; the tool does not label a population variance as a sample estimate.
Sources and further reading
These references support the documented method or context. They do not imply application testing or endorsement.